The fate and contamination of trace metals in soils exposed to a railroad used by Diesel Multiple Units: Assessment of the railroad contribution with multi-tool source tracking
Jacek Szmaglińskia,1, Nicole Nawrota,⁎,1,Ksenia Pazdrob, Jolanta Walkusz-Miotkb, Ewa Wojciechowskaa
aGdańsk University of Technology, Faculty of Civil and Environmental Engineering, Narutowicza 11/12, 80-233 Gdańsk, Poland
bInstitute of Oceanology of the Polish Academy of Sciences, Marine Geotoxicology Laboratory, Powstańców Warszawy 55, 81-712 Sopot, Poland
H I G H L I G H T S
• Tracking connection between railroad operation and trace metals in soils
• Statistical analyses and Pb isotope ratios applied as tools
• Cr supplementation associated with the specificity of railroad geometrical layout
• Coal combustion as a source of Pb and correlated metals: Zn, Cu, Cd
• Lower environmental footprint exhib- ited by railway than road transport
G R A P H I C A L A B S T R A C T
a b s t r a c t a r t i c l e i n f o
Article history:
Received 25 April 2021
Received in revised form 22 July 2021 Accepted 23 July 2021
Available online 29 July 2021
Editor: Filip M.G. Tack
Keywords:
Trace metals Railroad Pollution indices Pb isotopic ratio Soil
Diesel Multiple Units
Soil samples from cut slopes from lightly loaded railway lines used by Diesel Multiple Units for 5 years in Gdansk (Poland) were collected and analyzed for trace metals (TMs): Zn, Pb, Cd, Ni, Cr, Cu, and Fe. The main aim was to assess soil enrichment, contamination status, and distribution of TMs relative to the distance from the railway track. Extensive source tracking analyses were performed using cluster analysis (CA) and the Pb isotope ratios approach (206Pb,207Pb, and208Pb). Soil samples were affected by Cr, Cu, Pb, and Zn (max values in mg/kg d.w.:
31.1, 145, 80.5, and 115, respectively). The Enrichment Factor showed moderate (Cr, Zn, Pb) to very severe (Cu) enrichment. CA allowed TMs to be divided into two general groups: a) containing Zn, Pb, Cd with slight in- teraction with Cu; and b) containing Fe and Ni with slight interaction with Cr. Correlation analyses indicated Cr as an outlying TM delivered from a separate source associated with the specificity of the construction of railroad 248, where alloys containing Cr were used to counteract increasing wear-and-tear of the rails. Pb isotopic ratios in the ranges of 1.16–1.20 (206Pb/207Pb) and 2.05–2.10 (208Pb/206Pb) corresponded to anthropogenic supplemen- tation (coal combustion, road vehicles, and railroad transport) of Pb and Pb-correlated TMs (Zn, Cd, and partly Cu). Despite the research focus on the impact of the railroad contribution, a comparison with other forms of transport indicated that road transport appeared to have a higher contributing factor to TM pollution at the in- vestigated site. This general conclusion again emphasizes the lower environmental footprint exhibited by railway transport in comparison to road transport.
© 2018 Elsevier B.V. All rights reserved.
⁎ Corresponding author.
E-mail address:nicnawro@pg.edu.pl(N. Nawrot).
1Jacek Szmagliński and Nicole Nawrot contributed equally to this work.
https://doi.org/10.1016/j.scitotenv.2021.149300 0048-9697/© 2018 Elsevier B.V. All rights reserved.
Contents lists available atScienceDirect
Science of the Total Environment
j o u r n a l h o m e p a g e :w w w . e l s e v i e r . c o m / l o c a t e / s c i t o t e n v
1. Introduction
Globalization and economic growth in recent decades have led to dynamic development of transport. Transportation not only has a direct negative contribution to human health and life due to the risk of acci- dents (Jamroz et al., 2019;Kustra et al., 2019;Budzynski et al., 2019) but it also has a high environmental footprint, and it is beyond any doubt one of the major pathways for pollutants to enter air, water and soil (Yan et al., 2018). Pollution is associated with the combustion of liq- uid fuels, dust emission related to the wear of mechanical elements of vehicles and roads, followed by secondary dust emission related to par- ticles displaced by air movement caused by a passing vehicle (Adamiec et al., 2016;Crosby et al., 2014;German and Svensson, 2018;Liu et al., 2018b;Nawrot et al., 2020a). Transport is, therefore, a source of both point emission (production plants and refineries) and linear emission (roads, railways, etc.). The latter affects areas in the vicinity of traffic corridors and, typically for no-point emissions, is usually difficult to as- sess and properly manage. Exhaust gases and the effects of vehicle and road wear are a source of harmful substances such as oxides of carbon, nitrogen, and sulphur (from 10 to 40% of total emissions), suspended dust, trace metals (TMs) (Mohsen et al., 2018;Wojciechowska et al., 2019a,b), polycyclic aromatic hydrocarbons (PAHs) (Liu et al., 2020), polychlorinated biphenyls (PCBs) (Stojic et al., 2017), etc. Emission magnitude depends strongly on the means of transport. Currently, in Europe, about 75% of goods are transported by road, while approxi- mately 19% are transported by rail, and the remainder is transported by water (EC EUROPA). In the case of passenger transport, private road transport constitutes 62–96% and rail transport 0–17%, depending on the country (EEA EUROPA). Due to the clear disproportion in passenger-kilometers and ton-kilometers, as well as the emissivity of each means of transport, for many years research has focused on the im- pact of road transport (Chen et al., 2010;Hong et al., 2018;Zafra et al., 2017). Some other studies examined air transport, where toxic com- pounds are mostly released in the vicinity of runways (Brtnický et al., 2020). Research concerning the pathways for contaminants originating from rail transport have most often focused on dust emissions from the contact of wheels with rails, the wear of the overhead lines and slide plates of pantographs (Weerakkody et al., 2017), as well as the content of dust in the atmosphere of closed systems, such as metro or under- ground rail systems (Nieuwenhuijsen et al., 2007). Contaminants emit- ted from rail transport can cause health problems, which have been investigated by, among others,Loxham and Nieuwenhuijsen (2019).
Some earlier studies concerning TMs in urban areas selected Cr as a characteristic component in the bottom sediments of retention tanks located in the vicinity of railroads (Nawrot et al., 2020b).
The particularly important toxic elements deposited in the soils of railway cut slopes are TMs. Zn, Pb, Cr, Cd, Ni, and Cu which are of re- search interest in areas near railroads (Brtnický et al., 2020). TM occur- rence in the environment stamps a specific footprint due to their non- biodegradable nature. TM enrichment should be evaluated in the con- text of their natural persistence in the soil, for instance by using calcula- tion indices like the Pollution Load Index (PLI), Contamination Factor (CF), Enrichment Factor (EF), or Geochemical Index (Igeo) (Kowalska et al., 2018;Weissmannová and Pavlovský, 2017;Wojciechowska et al., 2019a,b). Scanning for TM enrichment in soils is an effective tool that accurately reflects the existing contamination of the surrounding environment and should be of interest to decision makers. Statistical methods to obtain data for different matrices are frequently used to ver- ify the possible sources of contamination (González-Macías et al., 2014;
Wang et al., 2015). An advanced method for source tracking of TMs in soil relies on the use of stable isotopes, which provide a“signature” of elements (Gao et al., 2018). In environmental science, the isotope ratios
206Pb/204Pb,206Pb/207Pb,208Pb/206Pb and208Pb/207Pb offer the most useful and reliable information due to their abundance and precise an- alytical determination. The Pb isotope signature in surface soil samples yields some basic information on Pb's origin, e.g. whether it is from
coal-fired power plants, burning of fossil fuels, waste incineration, min- ing, etc. (Cheema et al., 2020). The use of Pb isotope analysis can also in- directly point out the sources of other TMs since the abundance of various metals can be linked (Nawrot et al., 2020a). The application of Pb isotopes to track Pb and associated TM origin is a relatively novel ap- proach and, to the authors' knowledge, has not been applied before to examine the impact of railroad transport on soil.
Combining methods of contamination assessment and source track- ing of TMs benefits an authoritative assessment of soils. This approach supports decision-making and good management practices that might prevent environmental deterioration.
The overarching aim of this study is to assess the impact of railroad on the fate of TMs and contamination in soils at a lightly loaded railway line with cut slopes, which was built between 2013 and 2015. The rail- way is used by diesel trains. The specific research objectives were as fol- lows: 1) to assess TM (Pb, Zn, Cd, Cr, Ni, Cu, and Fe) enrichment and distribution in the surface soil layer of railway embankments relative to the distance from the railway track; 2) to evaluate the contamination status of surrounding soils including the differences between the natu- ral and anthropogenic levels of TMs of an analyzed area; and 3) to track the TM sources by employing statistical and Pb isotopic ratio analyses (206Pb/207Pb,208Pb/206Pb, and207Pb/208Pb). Thefindings of this study contribute to an indication of anthropogenic TM sources and the assess- ment of the overall impact of transportation modes on the environment.
2. Materials and methods
2.1. Study area and specific feature of the testing section
The railroad section tested in this study is located along railway line no. 248 in Gdansk, northern Poland. Railway line 248 was built between 2013 and 2015. Due to the challenging terrain, railway line 248 has the parameters of a mountain line (very large longitudinal gradients and small radius of curves). Currently, 46 pairs of trains run daily on railway line 248 in the research area. The trains consist of Diesel Multiple Units (DMUs) weighing from 82 to 108 t (108 Mg). Hence, the average daily truckload is 4600 t (4600 Mg), which gives an annual load of around 1.5 Tg. This classifies the railway line as very lightly loaded. The maxi- mum speed of the trains is 120 km/h. Railway line 248, following regu- lations enforced in Poland (Regulation of the Minister, 1998), is classified as Class 1 (on a four-level scale, where 0 is the highest class and 3 is the lowest). Since opening, the number of transported passen- gers has been gradually increasing and amounted to 400,000 people per month at the end of 2019. Electrification of the railway line will start in 2021, and the completion of work is planned for 2023. After electrifica- tion, the trains are expected to be faster and run more frequently, which should attract even more travelers.
2.2. Soil sampling strategy
Three testing sections (Fig. 1) were selected to extract soil samples in the close vicinity of the railway track. The boundary of the track is a ballast structure and the track itself was made as a continuous welded track, which consists of rails with 49E1 profiles, according to European standard EN 13674-1 on prestressed concrete sleepers of the PS-93/
W-14/1435/49E1 type with a W-14 fastening system.
The parameters of the testing sites presented inFig. 1are as follows:
• Test site 1 at 4 + 000 km (Fig. S. 1a): a double-track section located on a horizontal curve with a radius of about 800 m and a longitudinal slope of 18‰. Due to the very small radius of the curve hardened rails made from heat-treated steel were used here. The cross-section is located near a road with very heavy traffic (the traffic flow is more than 5000 vehicles per hour in the peak) (Via Vistula, 2016).
Three soil sampling points were located at this site: at the base of the embankment (native soil)– sample no. 11; at the half-height of
J. Szmagliński, N. Nawrot, K. Pazdro et al. Science of the Total Environment 798 (2021) 149300
2
Downloaded from mostwiedzy.pl
Fig. 1. Location of the testing sites. Figure elaborated using Google maps and OpenStreetMap.
Downloaded from mostwiedzy.pl
the embankment (embankment soil)– sample no. 12; and at the crest (sand and gravel mix 0–31.5 mm aggregate) – sample no. 13. This test site was chosen to verify the possible influence of the neighbouring in- tersection on metal soil supplementation on railway embankments.
• Test site 2 at 4 + 350 km (Fig. S. 1b): a double-track section located on a horizontal curve with a radius of about 800 m and a longitudinal slope of 18‰. Here, again, hardened rails were used due to the very small radius of the curve. The cross-section is located in an undevel- oped area. Seven soil sampling points were selected, of which a total of six were: at the base of the embankment (native soil)– sample no. 21; at the crest (sand and gravel mix 0-31.5 mm aggregate)– sam- ples no. 22 and 23; at the bottom of the drainage ditch (dry, native soil)– sample no. 24; and at the excavation slope (native soil) – sam- ples no. 25 and 26. One soil sample was extracted from the sand trap located at 4 + 640 km– sample no. 27 (not shown in Fig. S 1b).
• Test site 3 at 6 + 600 km (Fig. S 1c): a double-track section located on a horizontal curve with a radius of about 3000 m and a longitudinal slope of 13‰. Due to the radius of the curve typical rails were used there. The cross-section is located in an undeveloped area. Nine soil sampling points were selected: at the base of the embankment (native soil)– samples no. 31 and 36; at the half-height of the embankment (embankment soil)– samples no. 32 and 35; at the crest (sand and gravel mix 0–31,5 mm aggregate) – samples no. 33 and 34; at a 10 m distance from the base of the embankment (native soil)– sample no. 37; and from the drainage ditch– samples no. 38 and 39.
The soil samples were collected by using a plastic sampler. At each sampling point, the top layer of soil (approx. 5 cm in depth) from 3 to 4 crosswise located sites (within the 0.25 m2area of the sampling point) were extracted and mixed. Each mixed soil sample was packed into a PE bag and immediately delivered to the laboratory. The sampling process took place in June 2020.
2.3. Chemical analyses
2.3.1. Determination of trace metals
After collection, the soil samples were homogenized, transferred to Petri dishes, and lyophilized. To determine the concentration of seven TMs (Zn, Cu, Pb, Cd, Ni, Cr, and Fe) a soil subsample of 0.5 g (0.001 g ac- curacy) was digested with HClO4, HF, and HCl (3:2:1; Suprapur) in Tef- lon bombs in an oven (140 °C), for 4 h, according to the procedure described byVallius and Leivuori (1999). The solution was then evapo- rated to dryness and 5 mL of concentrated Suprapur HNO3was added and evaporated. The dried residue was dissolved in 5 mL of Suprapur 0.1 M HNO3and placed in polyethylene tubes. Dilutions of ×10, ×100, and ×1000 were prepared and analyzed in an inductively coupled plasma mass spectrometer (ICP-MS; Perkin–Elmer ELAN 9000). The re- sults are presented as mg/kg d.w. (d.w.– dry weight). The measure- ments were replicated three times. Quality control was assured by analyzing a certified reference material and “blanks”, according to the same procedure. Recoveries were in the range of 92–103%, depending on the individual metals. The precision, given as Relative Standard Devi- ation, was in the range of 3–5%. The detection limits (LOD) of each ele- ment were calculated as Blank + 3·SD, where SD values were the standard deviations of the blank samples (n = 5). LODs were as follows:
Pb - 1.0 mg/kg, Zn -0.5 mg/kg, Cr - 1.5 mg/kg, Ni - 0.7 mg/kg, Cu - 0.3 mg/kg, and Cd - 0.1 mg/kg.
2.3.2. Pb isotopic ratios
Stable lead isotopes (206Pb,207Pb and208Pb) were measured using a Perkin–Elmer Sciex ELAN 9000 ICP-MS and the206Pb/207Pb,208Pb/206Pb,
208Pb/207Pb ratios were calculated. Together with all sample sets the re- peated digestion and analyses of the certified standard material NBS- 981 (n = 5) was performed to validate measurement accuracy. The re- sults were satisfactory and recovery was >98%. The mean206Pb/207Pb ratio for NBS-981 was 1.09413 ± 0.00407 (certified value = 1.09333)
and the mean208Pb/206Pb ratio was 2.14269 ± 0.00859 (certified value = 2.16810). Three blank samples (containing only chemicals) were measured with every ten samples..
LODs were as follows:206Pb - 0.007μg/dm3,207Pb - 0.007μg/dm3, and208Pb - 0.007μg/dm3. Mean RSD was calculated from a triple analy- sis (separate digestion and measurement of 3 parallel subsamples) of every third sample. It was equal to 0.17% for the206Pb/207Pb ratio and 0.53% for the208Pb/206Pb ratio.
2.4. Soil contamination status
A geochemical assessment of the soil environmental status was per- formed using several soil indices. A Contamination Factor (CF) was ap- plied as an index which provides information about a single metal's contribution to the contamination status (Hakanson, 1980;Tomlinson et al., 1980) according toformula (1).
CF¼ CSample
CBackground ð1Þ
where CSampleis the metal content in the analyzed soil sample and CBackgroundis the metal content in soil referred to as preindustrial (this value informed the environmental background status). The CBackground
values were established based on the provided by the Central Geological Database in Poland (GeoLOG application) for the surface layer of soil (mean in mg/kg d.w. for the examined area: Zn– 45, Cu – 4.2, Cd – 0.5, Pb– 16.5, Ni – 5.6, Cr – 5.6).
For a comprehensive evaluation of soil quality in accordance with all the analyzed TMs, the Pollution Load Index (PLI) was calculated accord- ing to formula (2) (Tomlinson et al., 1980; Weissmannová and Pavlovský, 2017).
PLI¼ ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi CF1∙CF2∙ . . . ∙CFn
pn
ð2Þ where CFnis the CF described above and n is the number of metals studied.
Moreover, to further analyze the obtained results, an Enrichment Factor (EF) was calculated. This index considers the normalization of TM concentration and enables the identification of those areas of the ex- amined soils in which increased amounts of anthropogenic chemical components have accumulated, regardless of their granulometry (Sakan et al., 2014).Formula (3)presents the computational approach for EF (Zhang et al., 2012).
EF¼
CSample
CBackground CrefSample CrefBackground
ð3Þ
where CSampleand CBackgroundare the same values as described above, while CrefSampleand CrefBackgroundrepresent the normalizer element con- centration in the analyzed sample and environmental background, re- spectively. The normalizer element was Fe due to its common abundance in the earth's crust. The CrefBackgroundvalue, established as a mean concentration of Fe (15,866 mg/kg d.w.), was used in calculations for all the analyzed samples (to take into account the possible influence of extraneous soil used in the formation of railway embankments).
The classes of soil quality described with the use of CF, PLI, and EF are presented in Table S.1.
2.5. Source tracking using Pb isotopic ratio
A two-isotope plot (206Pb/207Pb vs.208Pb/206Pb) was applied to dis- tinguish separate Pb sources. This approach allows for a more precise identification of separate Pb sources. The values of the206Pb/207Pb and
208Pb/206Pb ratios reported in the literature are presented inTable 1.
J. Szmagliński, N. Nawrot, K. Pazdro et al. Science of the Total Environment 798 (2021) 149300
4
Downloaded from mostwiedzy.pl
2.6. Statistical analyses
Non-parametric Spearman correlation analysis was used for vari- ables (TMs) to assess their interdependence. Correlations with p <
0.05 were considered to be significant. Ward's cluster analysis (CA) for the measured TMs was carried out to identify the relationship between seven elements and their probable sources. The data analysis was per- formed using STATISTICA 13 (StatSoft) software.
3. Results & discussion
3.1. Trace metal content in the soil
TM content (mean ± SD) in the analyzed samples is presented in Table 2. The general scenario was that TM concentration decreased in the following order: Fe > Zn > Pb > Cr > Cu > Ni > Cd. In comparison to the background TM concentrations in the area under consideration, the highest excess in the analyzed soil samples was noted for Cr, Cu, and Zn: 5.6, 5.5, and 2.2 times for Test site 1, 5.2, 6.3, and 1.3 times for Test site 2, and 4.2, 34.5, and 2.6 times for Test site 3, respectively.
Stojic et al. (2017), in research performed in Serbia, found increased concentrations of Zn and Cu near to a railroad (within 1 km) and
concluded that railways, in general, contribute these TMs to surround- ing soils. The Cu and Zn concentrations reported byStojic et al. (2017) ranged between 8.73–216 mg/kg d.w. and 12.69–191 mg/kg d.w., re- spectively. In our study, the maximum observed contents of these TMs were 145 mg/kg d.w. for Cu and 115 mg/kg d.w. for Zn. The railroad in- vestigated in our study is relatively new, compared to the 130-year-old railway in Serbia and also relatively lightly loaded, which could account for the differences in TM concentrations. An important observation from our research was that Pb concentrations were higher in comparison to other studies– the maximum values measured in this investigation were 65 mg/kg d.w. at Test site 1, 41.1 mg/kg d.w. at Test site 2, and 80.5 mg/kg d.w. at Test site 3, whileStojic et al. (2017)reported a max- imum Pb value of 36.4 mg/kg d.w. andZhang et al. (2012)reported a maximum Pb value of 41.8 mg/kg d.w. This higher Pb content may be associated with the specificity of the analyzed area, since the railway is located in a city district (only Test site 3 is further the city centre and is located in a forested area). In contrast, the above-mentioned studies analyzed soil samples collected close to railways crossing farmlands or natural landscape zones. However, the Ni and Cd concentrations in this study were at a relatively low level (3.28–13.2 mg/kg d.w. and 0.030–0.282 mg/kg d.w. for Ni and Cd, respectively), whileStojic et al. (2017)reported a Ni content ranging between 11.0 and 115 mg/kg d.w. and Cd content between 0.080 and 0.740 mg/kg Table 1
The Pb isotopic ratios (206Pb/207Pb and208Pb/206Pb) for different sources of Pb in the environment.
Pb isotope ratio Value Description of the source of Pb origin References
206Pb/207Pb 0.96–1.2 Anthropogenic (Sun et al., 2018)
>1.2 Natural
1.22 Old and uncontaminated polish rocks (Zaborska, 2014)
1.16–1.17 Leaded gasoline used in central and eastern Europe (Yao et al., 2015)
1.14–1.15 Unleaded gasoline 1.14–1.16 Diesel
1.10 Unleaded gasoline and diesel from Russia (Chrastný et al., 2018)
1.15–1.17 Mechanical wastes derived from vulcanisation, tyre balancers, tyre scraps, and wheel alignment systems (Nawrot et al., 2020a)
1.17–1.19 European coal (Komárek et al., 2008)
208Pb/206Pb <2.08 Natural/coal source (Zaborska and Zawierucha, 2016)
>2.08 Contribution of gasoline Pb
2.05–2.08 Mechanical wastes derived from vulcanisation, tyre balancers, tyre scraps, and wheel alignment systems (Nawrot et al., 2020a)
Table 2
Concentration of mean ± SD (n = 3) trace metals (TMs) (Cr, Ni, Cu, Cd, Pb, Zn, and Fe) in analyzed soil samples collected from the Testing sites 1–3 along railway line 248.
Sample Trace metal mean ± SD concentration [mg/kg d.w.]
Cr Ni Cu Cd Pb Zn Fe
Test site 1
11 20.5 ± 0.8 7.24 ± 0.29 22.9 ± 0.92 0.112 ± 0.004 65.0 ± 2.6 59.9 ± 2.4 12,945 ± 259
12 31.1 ± 1.2 6.03 ± 0.24 7.58 ± 0.30 0.100 ± 0.004 23.5 ± 0.9 35.2 ± 1.4 13,748 ± 275
13 26.8 ± 1.1 8.36 ± 0.33 20.2 ± 0.81 0.282 ± 0.011 60.6 ± 2.4 97.5 ± 3.9 20,082 ± 402
Test site 2
21 29.3 ± 1.2 8.32 ± 0.33 26.5 ± 1.1 0.122 ± 0.005 41.1 ± 1.6 58.4 ± 2.3 19,267 ± 385
22 22.5 ± 0.9 9.53 ± 0.38 6.62 ± 0.26 0.090 ± 0.004 32.6 ± 1.3 37.5 ± 1.5 21,473 ± 429
23 23.9 ± 1.0 7.34 ± 0.29 8.64 ± 0.35 0.092 ± 0.004 22.9 ± 0.9 36.8 ± 1.5 14,303 ± 286
24 22.1 ± 0.9 3.28 ± 0.13 4.58 ± 0.18 0.056 ± 0.002 13.8 ± 0.6 20.3 ± 0.8 14,189 ± 284
25 22.3 ± 0.9 3.71 ± 0.15 4.96 ± 0.20 0.074 ± 0.003 15.6 ± 0.6 22.1 ± 0.9 11,667 ± 233
26 17.8 ± 0.7 3.99 ± 0.16 4.57 ± 0.18 0.074 ± 0.003 17.6 ± 0.7 20.3 ± 0.8 10,270 ± 205
27 9.61 ± 0.38 7.04 ± 0.28 12.6 ± 0.51 0.120 ± 0.005 22.7 ± 0.9 56.8 ± 2.3 11,705 ± 234
Test site 3
31 21.6 ± 0.9 5.21 ± 0.21 145 ± 6 0.174 ± 0.007 64.1 ± 2.6 60.6 ± 2.4 15,170 ± 303
32 21.6 ± 0.9 8.24 ± 0.33 26.0 ± 1.0 0.274 ± 0.011 66.3 ± 2.7 88.8 ± 3.6 11,972 ± 239
33 19.0 ± 0.8 6.45 ± 0.26 11.5 ± 0.5 0.086 ± 0.003 38.0 ± 1.5 33.9 ± 1.4 11,919 ± 238
34 22.2 ± 0.9 8.67 ± 0.35 12.0 ± 0.5 0.092 ± 0.004 25.7 ± 1.0 52.7 ± 2.1 31,494 ± 630
35 23.5 ± 0.9 13.2 ± 0.53 34.3 ± 1.4 0.136 ± 0.005 80.5 ± 3.2 75.8 ± 3.0 19,476 ± 390
36 15.9 ± 0.6 5.17 ± 0.21 13.5 ± 0.5 0.120 ± 0.005 48.1 ± 1.9 115 ± 4.6 11,010 ± 220
37 23.7 ± 0.9 8.50 ± 0.34 21.4 ± 0.9 0.206 ± 0.008 57.6 ± 2.3 71.5 ± 2.9 14,881 ± 298
38 11.7 ± 0.5 7.86 ± 0.31 10.6 ± 0.4 0.068 ± 0.003 16.4 ± 0.7 34.7 ± 1.4 9428 ± 189
39 17.5 ± 0.7 7.62 ± 0.30 10.8 ± 0.4 0.030 ± 0.001 20.8 ± 0.8 21.5 ± 0.9 6989 ± 140
Background (in surface soils) (GeoLOG application) 5.60 5.60 4.20 0.500 16.5 45.0
Local background 15,866
Downloaded from mostwiedzy.pl
Table 3
Trace metal (TM) concentrations [mg/kg d.w.] in soil samples influenced by rail, road, and air transport across the world.
Site Description Concentration of TMs [mg/kg d.w.] References
Zn Cu Pb Cd Ni Cr
Railway transport Masovian Voivodeship,
Poland
Railway line crossing a wooded area; built in 1936 as an electrified line; an important role in the national railway system; railway serves electric trains and freight (diesel) trains.
77a 3.1d
32a 3.2d
45a 3.6d
0.97a 1.9d
37.8a 7.6d
11.9a 2.9d
Radziemska et al., 2020
Serbia Soil sampling at 0–0.5 km distance from a railroad built 130 years ago; section used for passenger and cargo traffic
12.69–191.4b 61.27a 0.48d
8.73–215.7b 41.46a 1.02d
12.91–36.44b 24.16a 0.30d
0.08–0.74b 0.38a 0.52d
14.16–115.5b 48.99a 1.35d
10.32–64.26b 36.46a 0.36d
Stojic et al., 2017
Qinghai–Tibet D: soil sampling at 2–150 m distance from a railroad;
relativelyflat terrain;
T: soil sampling at 2–200 m distance from the railroad;
uniform vegetation and soil type associated with the regional landscape
H: soil sampling at 2–200 m distance from the railroad;
dominant landscapes in the Tibetan plateau;
D: 41.1 ± 11.9c 0.55d T: 81.8 ± 7.6c 1.10d H: 78.8 ± 8.7c 1.06d
D: 14.8 ± 1.5c 0.65d T: 31.1 ± 1.8c 1.38d H: 24.1 ± 1.9c 1.07d
D: 21.4 ± 6.5c 0.82d T: 26.1 ± 2.7c 1.00d H: 25.6 ± 3.1c 0.98d
D: 0.13 ± 0.09c 1.37d T: 0.19 ± 0.05c 1.91d H: 0.20 ± 0.09c 2.07d
D: 16.2 ± 2.9c 0.69d T: 33.6 ± 2.6c 1.44d H: 32.8 ± 2.8c 1.40d
D: 31.5 ± 5.3c 0.52d T: 71.2 ± 4.1c 1.17d H: 70.6 ± 6.2c 1.16d
Zhang et al., 2012
Abisco, Northern Sweden
Soil sampling at 0–1280 m distance from a railroad; railway transporting iron ore through an otherwise undisturbed, unpolluted birch forest
43–68b 25–64b 60–230b Goth et al., 2019
Sichuan Province, China Soil samples collected on a cut slope affected by railway transportation for 5 years
137.2–168.9b 158.0 ± 9.4c 2.30d
15.6–19.1b 17.6 ± 0.9c 0.66d
25.8–87.1b 49.2 ± 16.5c 2.54d
0.67–4.27b 2.08 ± 0.94c 20.8d
20.3–64.3b 35.6 ± 13.8c 0.72d
Chen et al., 2014
Road transport
Europe Roadside soils; at the distance of 0–5 m to the road edge 110–380b 20–60b 30–200b 0.04–2.0b 17–48b 17–60b Werkenthin et al., 2014
Stockholm, Sweden Area near a street in central Stockholm 41a 7.8–118b
57.6a 12.8–171b
7.2a 1.5–15.4b
0.12a 0.015–0.32b
2.9a 0.26–7.5b
6.1a 1.13–18.3b
Johansson et al., 2008 Edinburg, United
Kingdom
Area near a two-lane road, with light traffic.
A: near curb, B: 1 m from the curb.
A: 213a 107–457b 1.99d B: 211a 99–460b 1.97d
A: 57a 22–122b 1.30d B: 79a 26–220b 1.80 d
A: 118a 25–621b 4.21d B: 35a 6–102b 1.25d
A: 1a 0–2b 1.00d B: 2a 1–4b 2.00d
A: 15a 6–33b 0.75d B: 9a 3–15b 0.45d
A: 16a 5–76b 2.00d B: 15a 6–29b 1.88d
Pal et al., 2010
Warsaw, Poland Areas near to a highway 70–100b
3.7d
0.7–0.9b 2.2d
3.7–3.7b 0.13d
29–32b 0.5d
Lisiak-Zielinska et al., 2021
Turin, Italy Soil samples at braking sites (BS) and acceleration sites (AC) BS: 90–560b AC: 170–300b
BS: 60–380b AC:100–490b
BS:40–100b AC: 30–170b
BS: 180–270b AC:180–500b
BS: 350–420b AC: 380–850b
Padoan et al., 2017
Air transport Runway (airport) Warsaw, Poland
Landing site 3.96–1581b
152 ± 239c 6.1d
2.58–41.3b 14.0 ± 9.19c 2.2d
22.3–473b 64.2 ± 66.9c 3.2d
3.67–32.6b 8.83 ± 4.77c 1.8d
Brtnický et al., 2020
Runway (airport) Delhi, India
Landing site 142 ± 56.9c
2.8d
28.1 ± 6.11c 2.2d
42.9 ± 13.9c 2.2d
2.07 ± 1.21c 25.9d
42.9 ± 5.43c 1.5d
120 ± 46.1c 1.9d
Ray et al., 2012
Explanation toa, b, c: TM concentrations were presented as average valuea, min–max valuesb, or mean ± SDcdepending on the results presented by individual authors; Explanation tod: Contamination Factor, calculated based on the background concentration of TM presented by individual authors.
J.Szmagliński,N.Nawrot,K.Pazdroetal.ScienceoftheTotalEnvironment798(2021)149300
6Downloaded from mostwiedzy.pl
d.w. In the case of Cr the values observed in this study (9.61–31.1 mg/kg d.w.) in general were in a similar range to those obtained byStojic et al.
(2017)andZhang et al. (2012): 10.3–64.2 mg/kg d.w. and 18.5–46.5 mg/kg d.w., respectively. According to the regulations set by the Polish Minister of the Environment (2016) for soil surface assessment, the concentrations observed in this study do not exceed the recom- mended limit values for terrain belonging to group IV (road and railroad lands) (these limit values are, in mg/kg d.w.: Cu– 600, Zn – 2000, Pb – 600, Cd– 15, Cr – 1000, Ni – 500).
3.2. Comparison of TM concentrations in soils with different types of transport
TM concentrations in soil samples obtained in this study were com- pared to other studies analyzing different means of transport: railway, road, and air transport (Table 3).Goth et al. (2019)reported increased Pb content (max 230 mg/kg d.w.) near a railway transporting iron ore through an otherwise undisturbed, unpolluted birch forest. Despite the similar period of use of the railroad studied in China (5 years) (Chen et al., 2014), results similar to this study were recorded only for Pb (up to 87 mg/kg d.w.). Zn, Cd, and Cr contents were higher than those observed in this study. TM supplementation to the soil in the vi- cinity of the railroad certainly does occur, although thefinal effect on soil pollution may depend on many factors, i.e. the type of fuel used, the land use around the railroad, its age as well as the initial TM content in soil at the analyzed site. TM contribution from other sources may also add to the overall concentration. In addition, the level of soil contamina- tion should be referred to the geochemical background or preindustrial content of each studied TM. Taking into account the CF, the highest excess over the background value was recorded for Ni in Poland (CF = 7.6) (Radziemska et al., 2020) and Cd in China (CF = 20.8) (Chen et al., 2014).
Comparing the TM content in our study to the results obtained for soils affected by road transport in European countries (Pal et al., 2010;
Werkenthin et al., 2014;Lisiak-Zielinska et al., 2021), it can be unequiv- ocally stated that, in the latter case, the TM concentrations were much
higher. It was also demonstrated byPadoan et al. (2017) that the increase in TM content in soil samples is related to emissions caused by rapid changes in movement (acceleration and braking).
In the case of air transport high TM content was observed for Zn and Pb in Poland (Brtnický et al., 2020) and Cr and Cd in India (Ray et al., 2012) at landing sites. In the assessment of air transport pollution, the highest exceedance of background concentrations was recorded for Zn in Poland (CF = 6.1) (Brtnický et al., 2020) and Cd in India (CF = 25.9) (Ray et al., 2012).
3.3. Consideration of the contamination status consideration of soil in the area of the railway
To gain a more detailed understanding of TM contamination, CF, PLI, and EF values were calculated for the obtained data and presented in Table 4.
The general observation coming from CF calculations indicates that at all test sites there is considerable Cr contamination on crests and em- bankments. Moreover, at Test site 1, considerable Cu and Pb contamina- tion was noted at the base of the embankment and the crest. At Test site 2 increased contamination status was observed for Cu in sample no. 21 located at the base of the embankment. In the case of Test site 3, on em- bankment locations (at the base and half-height of the embankment), increased Cu and Pb contamination was observed (especially in the case of Cu with very severe enrichment observed for samples no. 31, 32, and 35).
The PLI value gives a comprehensive assessment of the contamina- tion status, which suggests the influence of anthropogenic activities.
According toDung et al. (2013), a PLI between 1 and 10 refers to a pol- luted environment. In this study, the unity of PLI (PLI = 1) was exceeded up to 2.5 times (in sample no. 31– the base of the embank- ment), which reflects a deterioration of the soil quality. The values of CF and PLI for samples no. 27 (sand trap), no. 38, and no. 39 (drainage ditch) reflected only low to moderate deterioration of site quality, which indicates that a small amount of TMs were deposited in the
Table 4
The assessment of TM contamination status of soil samples using Contamination Factor (CF), Enrichment Factor (EF) and Pollution Load Index (PLI).
Sample CF EF
Cr Ni Cu Cd Pb Zn Cr Ni Cu Cd Pb Zn PLI
Test site 1
11 3.7 1.3 5.5 0.2 3.9 1.3 4.5 1.6 6.7 0.3 4.8 1.6 1.8
12 5.6 1.1 1.8 0.2 1.4 0.8 6.4 1.2 2.1 0.2 1.6 0.9 1.2
13 4.8 1.5 4.8 0.6 3.7 2.2 3.8 1.2 3.8 0.4 2.9 1.7 2.3
Test site 2
21 5.2 1.5 6.3 0.2 2.5 1.3 4.3 1.2 5.2 0.2 2.1 1.1 1.8
22 4.0 1.7 1.6 0.2 2.0 0.8 3.0 1.2 1.2 0.1 1.5 0.6 1.2
23 4.3 1.3 2.1 0.2 1.4 0.8 4.7 1.4 2.3 0.2 1.5 0.9 1.2
24 4.0 0.6 1.1 0.1 0.8 0.5 4.4 0.6 1.2 0.1 0.9 0.5 0.7
25 4.0 0.7 1.2 0.1 0.9 0.5 5.4 0.9 1.6 0.2 1.3 0.7 0.8
26 3.2 0.7 1.1 0.1 1.1 0.5 4.9 1.1 1.7 0.2 1.7 0.7 0.7
27 1.7 1.2 3.0 0.2 1.4 1.3 2.3 1.7 4.1 0.3 1.9 1.7 1.2
Test site 3
31 3.9 0.9 34.5 0.3 3.9 1.3 4.0 1.0 36.1 0.4 4.1 1.4 2.5
32 3.9 1.5 6.2 0.5 4.0 2.0 5.1 1.9 8.2 0.7 5.3 2.6 2.3
33 3.4 1.1 2.7 0.2 2.3 0.8 4.5 1.5 3.6 0.2 3.1 1.0 1.2
34 4.0 1.5 2.9 0.2 1.6 1.2 2.0 0.8 1.4 0.1 0.8 0.6 1.3
35 4.2 2.3 8.2 0.3 4.9 1.7 3.4 1.9 6.7 0.2 4.0 1.4 2.4
36 2.8 0.9 3.2 0.2 2.9 2.5 4.1 1.3 4.6 0.3 4.2 3.7 1.6
37 4.2 1.5 5.1 0.4 3.5 1.6 4.5 1.6 5.4 0.4 3.7 1.7 2.0
38 2.1 1.4 2.5 0.1 1.0 0.8 3.5 2.3 4.3 0.2 1.7 1.3 1.0
39 3.1 1.3 2.6 0.1 1.3 0.5 7.1 3.1 5.9 0.1 2.9 1.1 0.9
Colors table cells represent the soil quality described in Tab.S1
Downloaded from mostwiedzy.pl
sediment or that the redistribution of TMs involving surface runoff from embankments and railway construction is minor.
According to the EF classification (Table S.1), the anthropogenic con- tribution to the overall site contamination status was shown by the EF values for Cr, Cu, and Pb. For these TMs the EF values varied between 2.0–7.1, 1.2–36.1, and 0.8–5.3, respectively. The highest value was re- ported for Cu at Test site 3 (sample no. 31), indicating very severe en- richment. As shown inTable 4, the EF values of Ni, Cd, and Zn varied between 0.6–3.1, 0.1–0.7, and 0.5–3.7 respectively, with average values of 1.5, 0.3, and 1.3. In general the EF values for the analyzed TMs follow the descending order: Cu > Cr > Pb > Ni > Zn > Cd. Test site 1 was lo- cated close to a main road intersection of streets, and therefore the ele- vated Cu and Pb levels, when compared to Test site 2 (located 350 m away), could be explained by the impact of motor vehicles and traffic.
In general, road deposited sediments (RDS) produced via vehicle wear-and-tear are a significant source of TMs. For example,Hong et al.
(2018)reported the Cu and Pb content in RDS in the following ranges:
25.66–310.75 mg/kg and 15.61–220.35 mg/kg, respectively. In urban areas, the RDS could be set in motion by wind or air turbulence from traffic and transferred with dust to green areas, soil, as well as to plant leaves adjacent to the streets (Zafra et al., 2017).
Analysis of the CF and EF values clearly indicates enrichment of the analyzed soil samples with Cr, Cu, and Pb linked to anthropogenic activ- ities, and thus potential pathways of these TMs into the environment deserve special attention. Cr is frequently a component of alloy steel used for rails, which may explain its abundance. However, other studies of railway-affected soils reported lower Cr enrichment in relation to background concentration (lower CF index) (Table 4).
3.4. Properties of railroad 248's construction and operation in regard to TM emissions
Particle emissions related to the wear of rails, vehicle wheels, brake blocks and discs, pantograph slide plates, and overhead contact lines (on electrified lines) or exhaust emissions (on non-electrified lines) are the main sources of TM pollution resulting from the operation and use of railway tracks. Due to the transfer of very high loads, rails wear along the vertical plane. Types of rail and steel alloys in current use sig- nificantly reduce the rate of vertical rail wear. In the curvilinear sections, there is additional friction of the wheelflange against the side surface of the railhead, which results in lateral wear. The greater the curvature of the track, the greater the wear. It is assumed that arcs with a radius of more than 3500 m can be treated as“slow wearing”, and arcs with a ra- dius of less than 1000 m as“fast wearing” (Santa et al., 2016). An addi- tional parameter that affects the rate of lateral wear of the outer rail track is the cant deficiency, which depends on the curve radius, train speed, and the cant used (Pombo, 2012;Powell and Gräbe, 2017).
The commonly adopted practice to counteract intensified rail wear in curves is to use an appropriate rail steel alloy. According to the guide- lines adopted by railway line No. 248 (PKM– 08), on arcs with a radius of 800 m or less, rails with strength Rm≥ 1100 MPa strength alloyed or heat-treated steel should be used. Such alloys are R350 HT (heat-treated according to PN-EN 13674-1 + A1:2017-07) or B1000 (alloyed). The alloy composition of rails with increased hardness has a higher content of metals such as Cr and Mn compared to typical rails made of R260 steel (according to the PN-EN 13674-1 + A1:2017-07 standard). In the inves- tigated test sections, 49E1 rails made of heat-treated R350 HT steel (Test sections 1 and 2) and R260 steel (Test section 3) were used. The radius of the arch was 800 m (Test sections 1 and 2) and 3000 m (Test section 3). With an annual train traffic intensity of 1.5 Tg, it can be assumed that the rate of vertical rail wear is insignificant, while the side wear of the rails would be approximately 0.3 mm/year (Test section 1 and 2) and 0.05 mm/year (Test section 3).
Train wheels are made of ER7 steel (PN-EN 13260 + A1: 2011) or P54T steel (PN-ISO 1005-1: 2017-03). The vehicles running on the 248 railway line have monoblock wheelsets made of ER7 steel. The vertical
wear of the tread at this point is about 0.02 mm/1000 km, and the side wear of theflange is about 0.16 mm/1000 km (Muhamedsalih et al., 2019). The alloy composition of steels used in railways is shown in Table S.2.
In addition, parts of the braking system are subjected to heavy wear.
The intensity of brake wear and the emission of particulate matter depends on the type of brake system: in the case of block brakes, the cast iron inserts and treads will wear out (Olofsson, 2011), while in the case of disc brakes, the composite brake blocks and discs are subjected to wear (Abbasi et al., 2011). Trains equipped with disc brakes, regenerative braking systems (with supercapacitors), and retarders, are currently running on the analyzed railway line. During normal service, trains use friction braking only in thefinal stopping phase only; therefore, along selected sections (on the route between stops), the impact of the wear of disc elements and pads on soil contam- ination may be ignored.
Hence, it can be concluded that the increased content of Cr and Cu may result from the wear of railway and vehicle wheels. This conclusion is consistent with research results collected byAbbasi et al. (2013)on the emission of solid particles in rail transport systems. The above- mentioned study showed that the PM2.5 and PM10 dust particles (not originating from the exhaust gas) have an average concentration of 0.142μg/m3for Cr and 0.944μg/m3for Cu. For Fe (which is the basic alloying component of steel), high concentrations were found - on aver- age 30μg/m3. The last pollution source along railways, but not the least, is emission related to traction energy, which contributes Cu to the environment (Hai He et al., 1998). However, since electric power is not used, any source of Cu caused by the wearing of the current collector and contact wires can be excluded in the case of Railroad 248. Since the railway line in its current state has never been burdened with regular steam traction traffic (there have only been occasional trips made by historic steam locomotives), all pollution resulting from coal combus- tion comes from external emissions. The line is only now being prepared for electrification, so the source of pollution can solely be re- lated to diesel engines.
3.5. Tracking analyses of TM sources
3.5.1. Correlation analyses between TMs
Spearman's correlation coefficients for analyzed TM concentrations are presented inTable 5. Significantly positive correlations were found between the following elemental pairs: Ni-Pb, Ni-Fe, Cu-Pb, Cd-Pb, Cd- Zn, and Pb-Zn. These results indicated that Ni, Cu, Cd, and Zn concentra- tions positively correlated with Pb concentrations, and thus those metals were likely to originate from common sources. Both elevated concentrations and high CF values of Cu and Pb in most soil samples in- dicate potential anthropogenic origin. In addition,Chen et al. (2014) noted a similar anthropogenic source for the Cd-Pb pair. There was no correlation between Cr and any other TM concentrations. At the same time, both CFs and EFs calculated for Cr pointed to considerable and moderate contamination status, respectively. This could suggest supple- mentation of Cr from a different source than other metals, and railroad 248 seems to be the most likely.
CA results presented inFig. 2divide the analyzed TMs into two groups with the significant association: a) group “a” consisting of Zn, Pb, and Cd within a close distance (<0.3 of height) and b) group“b” con- taining Fe and Ni within approx. 0.4 height. Cu demonstrated a relation- ship with group“a”, while Cr did so with group “b”. A more significant correlation was related to a lower cluster distance (Zhang et al., 2012).
Moreover, elements within the same group are expected to exhibit a common anthropogenic or natural source (Khan et al., 2011). Zn, Pb, and Cd are widely regarded as indicator metals in contaminated soils near highways (Wang et al., 2011). However,Zhang et al. (2012) noted that Zn, Pb, and Cd could also be delivered to the soil from rail traffic. TMs belonging to group “a” are undoubtedly correlated with an- thropogenic activity. Group“b” could be associated with the natural
J. Szmagliński, N. Nawrot, K. Pazdro et al. Science of the Total Environment 798 (2021) 149300
8
Downloaded from mostwiedzy.pl
